SOCIAL IDENTITY COMPLEXITY, ETHNIC/RACIAL IDENTITY, AND PROSOCIAL
ATTITUDES IN PRE-ADOLESCENTS
A Thesis in
for the Degree of
ii
The thesis of Chelsea O. Mayo was reviewed and approved* by the following:
Martha E. Wadsworth Professor of Psychology Thesis Advisor
Jonathan E. Cook Assistant Professor of Psychology
Yo Jackson Professor of Psychology
Melvin M. Mark Professor of Psychology Head of the Department of Psychology
*Signatures are on file in the Graduate School
iii
ABSTRACT
This study investigated how ethnic/racial identity relate to social identity complexity in youth.
Social identity complexity is a measure of the perceived homogeneity or overlap of a person’s
social ingroups and is generally related to positive intergroup attitudes. A diverse sample of pre-
adolescents (N = 97, Mage = 11.28) completed measures of social identity complexity, pro-social
attitudes, ethnic/racial identity, and familial ethnic socialization. Contrary to theory, social
identity complexity was not directly associated with pro-social attitudes. However, familial
ethnic socialization moderated the relationship between social identity complexity and pro-social
attitudes.
iv
List of Figures
Figure 1: Familial Ethnic Socialization as moderator of the association of social identity
complexity and prosocial attitudes.………………………………………………………………15
vi
Table 1: Correlation Matrix for Social Identity Complexity, Prosocial Attitudes, Ethnic Identity,
Ethnic Socialization and Related Measures.……………………………………………………14
vii
Acknowledgments
This material is based upon work supported by the National Institute of Mental Health under
Award No. 1R21 MH107631-01. Any opinions, findings, and conclusions or recommendations
expressed in this publication are those of the author and do not necessarily reflect the views of
the National Institute of Mental Health.
1
Introduction
As children leave middle childhood and approach adolescence, they are increasingly
required to navigate diverse environments and form relationships both with those who are like
them and those who are not. Their ethnic and racial identity affirmation tends to increase during
the transition to adolescence at the same time that they are exposed to more sources of bias and
discrimination (Umaña-Taylor et al., 2014). Minority children especially have to balance their
marginalized identity with other ingroup associations that enable them to build relationships in a
mostly White society. When people consider their membership in multiple groups they develop
perceptions about the extent that these groups overlap, a social-psychological construct known as
social identity complexity (SIC, Roccas & Brewer, 2002). Greater social identity complexity is a
marker of increased tolerance and positive affect toward outgroups, and research has shown that
adults are capable of having both high SIC and high ingroup identification (Costabile & Austin,
2017). Little is known, however, about children’s ability to balance social identity complexity
with high ingroup identification. The goal of this study, therefore, is to begin to understand how
young adolescent children navigate this process of integrating a complex social identity with an
ingroup orientation of strong ethnic and racial identification.
The relationship between ethnic identification and minority children’s attitudes toward
exclusion has not been fully elucidated. Some research suggests that minority children, as
frequent victims of discrimination, understand the negative consequences of exclusion and
therefore may be less likely to perpetrate it and more likely to qualify exclusion (by race or other
dimensions) as wrong. Minority children also may rely less on race in choosing friendships or in
perceiving similarity between dyads (Hitti, Mulvey, & Killen, 2017). On the other hand, high
identification with their ethnic group for some ethnic minority children has been associated with
2
negative evaluation of an outgroup, and only those who demonstrated identification with a
superordinate identity (e.g. a national identity as American) held less biases (Guerra et al., 2010;
Jugert, Noack, & Rutland, 2011). While much research has indicated how priming superordinate
identities can promote positive intergroup peer relationships for minority children, at the same
time, numerous studies indicate the protective value of ethnic and racial identification and
socialization for minority youth, particularly for those who experience discrimination (C. S.
Brown & Chu, 2012; Rivas-Drake et al., 2014; Seaton, Upton, Gilbert, & Volpe, 2014). Given
the protective value of both these elements, it is important to understand how high ethnic
identification and positive outgroup attitudes can coexist to maximize individual and societal
level benefit for minority children. Social identity complexity may be this mechanism.
Social Identity Complexity
Social identity complexity allows us to understand an individual’s social inclusivity and
subsequent positive intergroup relations. Roccas and Brewer (2002) defined social identity
complexity as the perceived overlap of membership among an individual's ingroups. Those who
perceive less overlap between their ingroups are considered high in social identity complexity
because each ingroup has some members who do not belong to the person’s other ingroups (e.g.
only a few baseball players are also Black, only a few Black peers are also baseball players).
Such an individual perceives multiple, distinct ingroups, meaning their social identity is more
inclusive because any given peer need not have membership in both social groups (be both Black
and a baseball player) to be considered an ingroup member. Individuals low in complexity
perceive greater overlap or convergence between membership of ingroups (e.g. almost all
baseball players are also Black, almost all Black peers are also baseball players), essentially
perceiving one homogenous ingroup. They therefore hold more rigid ideas of who constitutes an
3
outgroup member, leading to a more exclusive social identity. The construct has been captured
effectively by asking participants to name social groups they identify with and then rate
bidirectional pairings of those groups for membership overlap (e.g. “How many seventh graders
are basketball players?”) on a 5-point scale from “almost all” to “hardly any.” More ratings close
to “hardly any” overlap indicate greater social identity complexity.
Studies conducted to date on social identity complexity have proposed some personal and
contextual correlates for adults and youth. In adults, SIC has been associated with personal
attributes, such as liberalism vs. conservatism and need for cognition, the diversity of social
experiences provided the individual, and situational factors such as stress and manipulated
ingroup threat (Grant & Hogg, 2012; K. P. Miller, Brewer, & Arbuckle, 2009; Roccas & Brewer,
2002). Two studies by Knifsend and Junoven (2013, 2014) have examined social identity
complexity in youth. Their first study took a cross-sectional and longitudinal approach,
examining SIC in a group of seventh graders (ages 12-14) and reassessing the group as eighth
graders. This sample was moderately diverse (47-48% European American) and from one school
of relatively high socioeconomic status (11% free-and-reduced lunch). These youth indicated
moderate levels of social identity complexity which was related to positive inter-group attitudes
in seventh and eighth grade, suggesting continuity in SIC across time. Their second study used
more ethnically diverse samples from four schools of varied racial makeup (19% to 48%
European-American) with a total sample that was 41% Latino/a or Mexican American and 34%
European American. These seventh graders also had moderately high levels of social identity
complexity. Additionally, for those with a high availability of cross-ethnic peers in their school,
SIC mediated the relationship between cross-ethnic friendships and ethnic intergroup attitudes.
Ethnic and Racial Identity Development
4
Other than the two pioneering studies by Knifsend and Junoven, very little research has
been done to understand the development of social identity complexity in youth. This is despite
the fact that pre-adolescence is a crucial time to examine SIC since children are spending less
and less time with the highly convergent “ingroup” of their family (where usually identities such
as race, religion, or SES naturally overlap), their peer relationships are becoming more important
(and cross-ethnic friendships usually decline), and they are cognitively maturing in their ability
to recognize ingroup and outgroup similarities and differences (Roccas & Brewer, 2002; Umaña-
Taylor et al., 2014). The few studies that have examined SIC in pre-adolescents considered the
context of school diversity and outcomes related to cross-ethnic friendships and ethnic intergroup
attitudes (Knifsend & Juvonen, 2013, 2014), however they did not consider the strength of
children’s ethnic/racial identification and how this could interact with social identity complexity.
Considering that ethnic/racial identity is the most common social group included in SIC
studies, and that SIC has been so often linked to tolerance of ethnic/racial outgroups specifically,
it is surprising that very few of these studies have considered the extent that a person identifies
with or affirms their ethnic/racial identity. This is perhaps because most studies of social identity
complexity have been conducted with majority White samples. The few studies that have
included diverse enough groups to make comparisons suggest that SIC’s relationship to ethnic
outgroup attitudes works similarly across racial groups (Brewer, Gonsalkorale, & van
Dommelen, 2012; Knifsend & Juvonen, 2013, 2014), but more studies with diverse samples are
greatly needed to understand how identifying with marginalized ingroups affects perceptions of
ingroup complexity.
Currently there are very few effective interventions for reducing discrimination and
racism, especially for children. Unfortunately, evidence shows that not only are biases formed at
5
a young age but without intervention they remain constant (Rubin, Bukowski, & Laursen, 2009).
Killen and Smetana (2010) call for interventions which consider the ways negative perceptions
of group identity and membership lead to bias and discrimination. Social identity complexity
may be a mechanism of group identification that mitigates or counteracts negative inter-group
processes. There has been some evidence that addressing the cognitive processes associated with
bias in children is a more effective way to reduce prejudice than mere contact with outgroups
(Aboud & Amato, 2001). Targeting SIC is potentially a more effective intervention against
prejudice than other methods, such as cursory cross-cultural exploration or white-centered anti-
racism teaching, which can be limited or even harmful (Aboud & Amato, 2001). Understanding
the development of social identity complexity at the pre-adolescent stage may lead to well-timed
interventions, ideally occurring before more exclusive mindsets (and subsequent bias, prejudice,
and discrimination) become entrenched in adolescence or adulthood.
While encouraging SIC can reduce tendencies toward discrimination in youth, it may also
be especially relevant to minority groups as a potential protective factor against ingroup threat.
Roccas and Brewer (2002) proposed that those who perceive their ingroups to be highly
overlapping (less complex identities) essentially perceive themselves as having one homogenous
ingroup, and therefore would be more likely to perceive an attack on any one ingroup as an
attack on their entire identity. Those who perceive their various ingroups as more distinct would
perceive ingroup threat as separate from their other sources of identity. This protective value of
social identity complexity could also be understood through the lens of multiculturalism. The
multiculturalist identity profile has been defined as a form of cross-racial identity which includes
both strong ethnic/racial identity and connectedness to other groups (Worrell, Andretta, &
Woodland, 2014). A strong dual identity may be more adaptive than identity profiles which
6
devalue either one’s own ethnic identity or positive relations with other ethnic groups.
Multiculturalism that includes self-acceptance and engagement with other cultural groups has
been related to positive clinical outcomes, academic achievement, and reduced vulnerability to
stereotype threat (Worrell et al., 2014). Social identity complexity may underlie the dually
adaptive function of multiculturalism, but few, if any, studies have examined social identity
complexity’s relationship to both outgroup and ingroup attitudes.
Emerging developmental theories of social identity position ethnic identity and prosocial
behavior as promotive factors for children exposed to risk (M. A. Zimmerman et al., 2013). In
order to understand and encourage the resiliency process in marginalized youth, it is important to
understand how their social identities integrate both self-affirmation and empathy toward others.
Social identity complexity has been shown to be a useful predictor of ethnic outgroup attitudes in
adults and pre-adolescents. However, studies have not adequately articulated SIC’s relationship
to one’s ethnic and racial ingroup attitudes. Strong identification with an ingroup is a double-
edged sword: belonging to social groups is essential and adaptive for an individual and society,
but it can also bolster comparisons against outgroups that lead to stereotyping and bias (Killen &
Smetana, 2010). Social identity complexity, however, is not about tolerance through a lack of
group identification, but rather through the recognition of multiple, distinct groups, enhancing an
individual’s ability to find common ground with different people in different ways. Social
identity complexity may be the key to both encouraging ethnic and racial identification without
maladaptive exclusion and improving outgroup inclusion without self-erasure.
The aim of this study was to examine whether social identity complexity distinguishes
those who affirm their racial/ethnic identity so much as to be exclusive of those outside their
group, from those who highly affirm their racial/ethnic identity but adopt more pluralistic and
7
tolerant mindsets. Therefore, high ethnic/racial identification was not predicted to be associated
with SIC positively or negatively alone, but rather SIC and ethnic identity affiliation were
conceptualized as interactive elements of social identity which shape pro-social attitudes. Based
on the literatures reviewed above, the study’s first hypothesis was that higher levels of social
identity complexity would be associated with greater prosocial attitudes in a diverse, majority-
minority (i.e. predominately non-White) sample of pre-adolescents, confirming generalizability
across racial groups of the association found in predominantly White samples. The second
hypothesis predicted moderation of this relationship by ethnic identity affirmation – specifically
that their positive association will be much stronger in the context of high ethnic affirmation. In
other words, SIC was expected to be especially important for positive outgroup attitudes when
there was also high ingroup affiliation (which might otherwise discourage outgroup tolerance).
A third hypothesis expected a similar moderation of the relationship between social
identity complexity and prosocial attitudes by familial ethnic socialization. Much of children’s
understanding of their ethnic identity comes through how it is socialized within their families
initially (Killen & Smetana, 2010; Umaña-Taylor et al., 2013). While generally strongly
correlated, ethnic identity may begin to diverge from familial ethnic socialization at this age
group as children become more susceptible to external influences and peer socialization (Umaña-
Taylor et al., 2013). Familial ethnic socialization may also tap into how social identity
complexity itself is socialized versus how it is expressed through ethnic affirmation, therefore it
was of interest to examine both variables. It was hypothesized that high ethnic socialization
would also engender a significantly stronger positive association between social identity
complexity and prosocial attitudes. SIC was expected to be particularly important for positive
outgroup attitudes when ingroup affiliation had been especially socialized.
8
Method
Participants
The study used a diverse sample of ninety-seven (n = 59 females, 63%) 10 to 12-year old
youth (Mage = 11.28) who completed interviews and questionnaires as part of an assessment for a
coping skills and identity development intervention for low-income pre-adolescents. The sample
of young adolescents identified as being Black or African-American (n = 58, 58.6%), Caucasian
or White (n = 15, 15.2%), Multiracial (n = 14, 14.1%), or other/did not report (n = 10, 10.1%).
Thirty-eight participants (38.4%) reported their ethnicity as Hispanic or Latino/a. The majority of
families were very low income, with an average annual household income of $19,742, and 68%
of families were receiving public assistance. Incomplete data due to missing items (ethnic
identity affirmation = 18 participants; prosocial attitudes = 3 participants) resulted in a lower n
for some of the analyses.
Procedures
Children were recruited from an urban city in central Pennsylvania through local
community and school events and contact with child-serving agencies. Parents and children were
initially phone-screened to determine eligibility. Families were required to have an annual
household income below 200% of the federal poverty level. Children with clinically elevated
anxiety or depression, or who were diagnosed with autism or intellectual disability were also
excluded. Eligible children and one parent or guardian were scheduled for a three-hour pretest
assessment during which they were consented for the study. Participants completed several
activities, interviews, and questionnaires during preassessment, only some of which are relevant
to the current study (see below). Parent-child dyads received $40 upon completion and were then
9
randomized to an intervention or control condition, though only preassessment data was utilized
in this study.
Social identity complexity. The Social Identity Complexity measure (Roccas & Brewer,
2002), was originally adapted for this age group by Knifsend and Juvonen (2013). For this study,
the measure was presented to the children as “Groups that Describe Me.” Children were read the
following prompt: “We would like to know which groups you belong to and which of those
groups are most important to you. Below are some examples of groups that could describe you.”
Participants were then given examples from the categories Gender, Race/Ethnicity, Religious
groups, Neighborhood, and School. Three example children of different races, genders, and
religious groups were presented and then the participants were asked to list their racial/ethnic
group, gender, religious group, and school affiliation.
Traditionally, studies of SIC have allowed participants to specify the social groups most
important to them and used those groups for the questions about overlap. This approach was
confusing to the pilot participants for this study, so we therefore followed the practice used in
Brewer, Gonsalkorale, and van Dommelen (2012) and pre-selected the four social groups for
them (race/ethnicity, gender, religion, and school). Participants were then asked to rate how
important each of those four group affiliations were to them on a 5-point Likert scale from
“Definitely Important” to “Definitely Not Important.” Finally, participants were asked to rate the
overlap between pairs of each of the four groups. They were first given practice prompts: “How
many students in the 7th grade are Basketball Players?” and “How many Basketball Players are
students in the 7th grade?,” were asked to rate each on a 5-point scale including “Almost All,”
“Most,” “Almost Half,” “A Few,” and “Hardly Any,” and then were asked to rate the
10
unidirectional overlap of their four groups the same way (a total of 12 pairings). An average
social identity score was computed for each participant based on their 12 ratings from the SIC
measure, whereby a higher average score (more “Hardly Any” overlap responses) indicated
greater social identity complexity.
Because each bidirectional pairing was ultimately asked two ways non-sequentially, the
pairings were each divided into two subsets (i.e. ‘How many Group A are Group B’ and in
another subset ‘How many Group B are Group A’) in order to estimate reliability. A Spearman-
Brown split-half reliability coefficient was calculated by computing averages for each subset and
correlating these composite scores. Adolescent and adult studies that have used this method to
calculate the reliability of the social identity complexity measure have found good internal
consistency with a reliability coefficient of .91 (Knifsend & Juvonen, 2013; Schmid, Hewstone,
Tausch, Cairns, & Hughes, 2009). The reliability coefficient for the current study’s pre-
adolescent social identity complexity measure was .86, indicating good internal consistency.
Importance of social identities. Social identity complexity theory presumes that each
ingroup has importance and meaning for that individual, so participants were also asked to rate
the importance of each group named (e.g. “How important is it that you are…”). This was
especially important in this study because the groups were pre-ascribed for participants, and the
centrality of race/ethnicity is being specifically targeted. Responses were on a 5-point scale (1 =
Definitely not important, 2 = Not very important, 3 = Somewhat important, 4 = Very important,
5 = Definitely important).
Ethnic identity. The Multigroup Ethnic Identity Measure (MEIM; Phinney, 1992) was
used to assess ethnic identity. The MEIM assesses ethnic identity as a two-factor structure of
ethnic identity search (exploration) and affirmation (belonging and commitment). The measure
11
includes 12-itmes: five items related to ethnic identity search and seven items related to
affirmation. This study focused on the affirmation items, such as “I have a lot of pride in my
ethnic group.” Participants were asked to rate each item on a five-point scale from “Strongly
Disagree” to “Strongly Agree.” An ethnic affirmation score was computed by averaging the
seven pertinent items from the Multigroup Ethnic Identity Measure. Studies using this measure
show good reliability, with Cronbach’s alpha over .80 for adolescent samples.
Familial ethnic socialization. Given the extent that ethnic identity in children is closely
tied to familial practices, the Familial Ethnic Socialization Measure (FESM; Umaña-Taylor,
Yazedjian, & Bámaca-Gómez, 2004) was included. The child was prompted to consider his/her
ethnicity and respond in relation to that identity. This is a 12-item scale that uses statements such
as “My FAMILY teaches me about my ethnic/cultural background.” Participants responded on a
five-point scale from “Not at all true” to “Very much true.” FESM has a reported reliability of
.92 (Umaña-Taylor et al., 2004). The ethnic socialization scores were computed by averaging all
items from the Familial Ethnic Socialization Measure.
Prosocial attitudes. Participants completed the Short Form Positive Youth Development
(PYD) inventory (Lerner et al., 2005), a 34 item measure composed of five “C” latent factors;
“Confidence,” “Competence,” “Character,” “Caring,” and “Connection.” For this study, 12 items
were selected to represent prosocial attitudes. Six items from the “Character” factor, specifically
all items from the social conscience, personal values, and valuing of diversity subscales, were
included, as well as all six items from the “Caring” factor which is meant to demonstrate
empathetic concern (Geldhof, Bowers, Boyd, et al., 2014; Geldhof, Bowers, Mueller, et al.,
2014). These items were assessed with 5-point scales with slightly varying prompts; for
example, items from the social conscience subscale were prompted as “How important is each of
12
the following to you in your life?” and items from the valuing of diversity subscale were
prompted “Think about the people who know you well. How do you think they would rate you
on each of these?” Generally average scores are computed within each subscale and factor, but
for this study a composite prosocial attitude score was computed by averaging the 12 target items
in addition to using the average score of the valuing diversity subscale.
13
Results
Preliminary Analyses
Participants showed on average a moderate level of social identity complexity, reporting
that between “Most” and “Almost Half” of the members of one ingroup also belonged to their
other ingroups (M = 2.86, SD = .78). Prosocial attitudes were generally elevated with our
participants, rating prosocial feelings and behaviors as highly important to them and indicating
that others would rate them highly for valuing diversity (M = 4.09, SD = .76). They also rated
each social group on the SIC measure to be moderately to highly important (M = 3.85, SD =
1.11, range = 3.53 – 4.12, on a 5-point scale), with gender typically being rated as most
important (M = 4.12, SD = 1.05). Racial and ethnic identity was rated on average between
“Somewhat Important” and “Very Important” to our participants (M = 3.83, SD = 1.12).
Several analysis of variance (ANOVA) models were used to explore group differences by
age, gender, race, and ethnicity (Hispanic or not) in social identity complexity, prosocial
attitudes, ethnic identity affirmation, and familial ethnic socialization. Participants did not
significantly differ on target variables by age, gender, race, or ethnicity so these were not
included as covariates in subsequent regression models, allowing the initial hypothesis to be
tested with a bivariate correlation.
Primary Analyses
As evident in Table 1, ethnic identity affirmation and familial ethnic socialization were
strongly positively correlated (r = .64, p < .001). Affirmation and socialization were also
moderately associated with importance of racial/ethnic identity as asked in the social identity
complexity measure (r = .39, p = .001 and r = .49, p < .001, respectively). The bivariate
correlation of social identity complexity and prosocial attitudes however was non-significant. In
14
order to examine interethnic attitudes more specifically, correlations were computed using the
average of just the two items on the PYD relating to valuing diversity. The association between
social identity complexity and these diversity items was also non-significant.
Table 1 Correlation Matrix for Social Identity Complexity, Prosocial Attitudes, Ethnic Identity, Ethnic Socialization and Related Measures
Variable 1 2 3 4 5 6 1. Social Identity Complexity — -.15 -.01 -.19 -.22* -.26** 2. Prosocial Attitudes -.15 — .64** .58** .36** .33** 3. Attitudes toward diversity -.01 .64** — .45** .35** .28** 4. Ethnic Identity Affirmation -.19 .58** .45** — .64** .39** 5. Ethnic Socialization -.22* .36** .35** .64** — .49** 6. Importance of Race/Ethnicity -.26** .33** .28** .39** .49** — p < .05*, p < .01**
In fact, prosocial attitudes, as measured by the PYD, were significantly correlated with all
variables except social identity complexity. Greater scores for ethnic identity affirmation,
familial ethnic socialization, and importance of race/ethnicity all predicted greater prosocial
attitudes as well as the subset of items about valuing diversity. Among the main variables of
interest, social identity complexity was only significantly correlated with Familial Ethnic
Socialization (r = -.22, p = .032) and the importance of race/ethnicity as reported on the SIC
measure (r = -.26, p = .009), and both were negatively associations. Thus, the hypothesis that
greater social identity complexity would be associated with greater prosocial attitudes was not
supported.
A moderation analysis was conducted to test for an interaction of ethnic affirmation and
social identity complexity, with the hypothesis that ethnic identity affirmation would moderate
the relationship between social identity complexity and prosocial attitudes. A multiple regression
model was run which included social identity complexity, ethnic identity affirmation, and their
15
interaction as predictors of prosocial attitudes. Results of this regression model indicated no
significant interaction of social identity complexity and ethnic affirmation.
A second similar moderation analysis was conducted to test whether the relationship
between social identity complexity and prosocial attitudes depends on ethnic socialization. It was
hypothesized that the relationship between social identity complexity and prosocial attitudes
would be strong and positive when ethnic socialization is high, and weaker when ethnic
socialization is low. A multiple regression model was run which included social identity
complexity, ethnic socialization, and their interaction as predictors of prosocial attitudes. Results
indicated a significant interaction of social identity complexity and ethnic socialization above
and beyond the effects of the two predictors, b = .17, η2 = .05, p = .028, 95% CI [0.02, 0.32]. As
shown in Figure 1, social identity complexity was positively associated with prosocial attitudes
when ethnic socialization was one standard deviation above its mean but negatively predicted
prosocial attitudes when ethnic socialization was one standard deviation below its mean.
16
Discussion
Building on preliminary developmental work using social identity complexity as a
measure of perceived ingroup homogeneity and outgroup attitudes, this study incorporated
ethnic/racial identity and familial ethnic socialization to examine how high ingroup identification
manifests with prosocial and tolerant attitudes in pre-adolescents. Ultimately, the study did not
find a relationship between social identity complexity and prosocial attitudes, nor did it find any
moderation by ethnic identity affirmation. However, greater familial ethnic socialization was
itself related to less social identity complexity, and the relationship between social identity
complexity and prosocial attitudes was moderated by ethnic socialization. In the context of high
ethnic socialization, greater social identity complexity did predict greater prosocial attitudes, but
in the context of low ethnic socialization, greater social identity complexity predicted less
prosocial attitudes.
Social identity complexity was not associated with prosocial attitudes in this study. These
findings contradict the core of social identity complexity theory, so it is important to consider
possible alternative explanations. First, this appears to be the youngest sample of participants on
which social identity complexity has been examined. It may be that the cognitive skills needed
for multiple, simultaneous classification are still maturing in pre-adolescents, making even the
subjective perception of ingroup overlap more tenuous (Aboud, 1989; Bigler & Liben, 1992;
Pfeifer, Spears Brown, & Juvonen, 2007). Assessment of this construct proved difficult initially
with pilot participants, hence the choice to use pre-ascribed groups. It is possible that even with
this adjusted format the pre-adolescents still found the task confusing, providing an inaccurate
representation of their overlap perceptions, despite good internal reliability of the measure.
17
It is also worth considering the skewed responses to the measure of prosocial attitudes.
The Positive Youth Development scale evidenced a significant ceiling effect with our
participants, such that in general the participants rated themselves highly for social conscience,
caring, and valuing diversity. This age group was found to score highly on the PYD in the 4-H
Study of Positive Youth Development from which the measure was derived. In that sample,
mean PYD scores for the 10 to 13 year-olds surveyed were well above scale midpoints (S. M.
Zimmerman, Phelps, & Lerner, 2008). In fact, the individual items targeted in the present study
were all moderately negatively skewed in the wave of sixth graders from the 4-H study (Geldhof,
Bowers, Boyd, et al., 2014). Future social identity complexity work in this age group could
incorporate prosocial attitude scales with higher ceilings. It is also possible that the participants
were motivated to respond in socially desirable ways on the PYD because the questions were
about overtly prosocial beliefs, a phenomenon that is generally common in measures of attitudes
(Krumpal, 2013). Our participants answered questions in the presence of an interviewer, which
has been shown to increase the likelihood of social desirability bias in children of similar age (P.
H. Miller et al., 2015), however this study did not include a method to measure or control for this
possibility. Though studies of social identity complexity to date have relied on self-reported
intergroup attitudes, including the studies with youth, future research could examine whether
external reporters of children’s prosocial attitudes or behavior (such as parent report) would
provide more accurate assessment of the effect of social identity complexity.
Furthermore, the construct of prosocial attitudes as a dependent variable may also be
meaningfully different from the construct of intergroup attitudes used in other studies of SIC.
The PYD items which assess prosocial attitudes, such as “When I see someone being taken
advantage of, I want to help them,” may be too broad to tap into the same mechanism of SIC that
18
relates to tolerance of outgroups. Generally, social identity complexity studies have assessed
intergroup attitudes with more explicit measures of social distance, for example using a ‘feelings
thermometer’ to ask how warm or cold participants feel toward specific ethnic/racial groups
(Brewer & Pierce, 2005; K. P. Miller et al., 2009; Schmid et al., 2009) or by assessing attitudes
toward race-related political issues. Knifsend and Junoven (2014) incorporated behavioral items
of outgroup distance designed to be developmentally relevant to their twelve to fourteen year-old
participants, for instance ratings of how much the participant would like to eat lunch with or sit
next to on the school bus with youth of four different ethnic/racial groups. In their earlier study
(2013), they used the other-group orientation subscale of the original Multigroup Ethnic Identity
Measure (MEIM) to assess openness to interaction with other groups, a subscale which is no
longer included in the updated MEIM version used in this study (Phinney, 1992). They also used
additional measures of perceived risks and opportunities of cross-ethnic interactions (Knifsend &
Juvonen, 2014). These types of items may be more similar to the two race-related items of the
PYD measure in this study, which asked participants to rate themselves for “Knowing a lot about
people of other races” and “Enjoying being with people who are of a different race than I am.”
However, the valuing diversity subscale was not predicted by SIC scores either.
Social identity complexity has been mostly examined in primarily European-American
samples, compared to this study’s sample of which only 15% of participants identified as White.
While SIC was fairly normally distributed in our sample, SIC levels did not differ by race or
ethnicity. It is therefore not obvious what conditions gave rise to more complex social identities
in some participants as compared to others. Some research suggests that contextual diversity may
promote social identity complexity (K. P. Miller et al., 2009). Though the sample as a whole for
this study was particularly diverse and our racial and ethnic distributions are similar to those of
19
the schools from which we recruited for the study (e.g., majority Black schools with 22-24%
Hispanic and 3-19% White), we did not assess for the availability of cross-ethnic peers or
neighborhood diversity on an individual level. Knifsend and Junoven (2014) chose four schools
that were moderately diverse for one of their studies with youth and they did find some
differences in SIC between racial/ethnic groups. The majority ethnic groups in each school
(usually Latinos) had less social identity complexity compared to the minority groups (Asians),
which they explained by the inherently different probabilities of cross-ethnic contact. However,
with every racial group the relationship between social identity complexity and outgroup
tolerance was still present and positive, and the availability of cross-ethnic peers was not directly
associated with SIC. Brewer and Pierce (2005) also found that demographically heterogenous
local environments were not necessarily an essential condition for individuals to demonstrate
high social identity complexity. Further research on social identity complexity in majority-
minority samples and more comparative studies are needed to understand how social identity
complexity might function differently in dissimilar demographic contexts.
The second hypothesis regarding an interaction of ethnic identity affirmation and social
identity complexity was also not supported. Given that ethnic identity affirmation was found to
be strongly correlated with prosocial attitudes, ethnic identity affirmation may be a promotive
factor for minority pre-adolescents independent of social identity complexity. In a majority-
minority environment, prosocial attitudes may not reflect outgroup attitudes for minority children
because they are more likely to encounter and interact with others of the same minority status (a
perceived ingroup). Future studies should better distinguish interethnic group attitudes from
general prosocial attitudes when considering the strength of ethnic identity and social identity
complexity.
20
The third hypothesis regarding an interaction of familial ethnic socialization and social
identity complexity was partially supported. It was hypothesized that the relationship between
social identity complexity and prosocial attitudes would be stronger when ethnic socialization
was high, and non-existent or weaker when ethnic socialization was low – based on the
expectation that SIC would be positively associated with prosocial attitudes. In this sample, the
relationship between SIC and PYD was positive in the context of high FESM but was negative in
the context of low FESM. This suggests that the optimal context for prosocial attitudes is in fact
the combination of high ethnic socialization and high social identity complexity, but perhaps a
non-overlapping (more complex) identity is detrimental when there has been less ingroup
socialization by family members. It may be that perceiving distinct ingroups without having been
socialized to affirm a core ingroup – especially an ingroup such as race/ethnicity that is more
likely to be treated as an outgroup by society – creates a similarly apathetic attitude toward one’s
other potential ingroups and therefore minimal prosocial attitudes. Future work should examine
socialization practices that encourage the perception of different ingroups but discourage the
attribution of value to those differences, and whether this type of socialization is detrimental to
personal identity formation or intergroup relations. For this population, social identity
complexity may not be essential for prosocial or tolerant attitudes, while ethnic identity
affirmation and familial ethnic socialization may be sufficient.
An additional limitation of this study is its use of cross-sectional data, preventing
conclusions about social identity complexity, ethnic identity affirmation, or familial ethnic
socialization causing prosocial attitudes. Knifsend and Junoven’s (2014) longitudinal study did
find that higher social identity complexity in seventh grade predicted ethnic intergroup attitudes
in eighth grade, but this relationship became non-significant once they controlled for seventh
21
attitudes or tolerance are therefore crucial to understanding how social identity complexity
develops over time and what factors are associated with its growth and change. In addition, such
data are needed to determine if it is possible to encourage SIC through intervention and if doing
so would have positive effects.
Though this study suggests social identity complexity is not always tied to prosocial or
interethnic attitudes, research is needed to determine if SIC may have protective value against
discrimination. Self-complexity theory suggests that identifying with multiple separate ingroups
buffers against threat to any one ingroup, as the individual’s identity is not solely staked in that
one group (Evans, 1994; Linville, 1985). Following this theory, those with highly overlapping
ingroups could be more likely to perceive an attack on one ingroup as an attack on their entire
identity, which could contribute to psychological maladjustment. Longitudinal research on SIC
and clinical outcomes could better evaluate SIC as a promotive or protective factor, in a similar
way that ethnic identity and familial ethnic socialization have been found to be both promotive
and protective against discrimination (M. A. Zimmerman et al., 2013).
Conclusions
The current study sought to expand research on social identity complexity, or the
convergence of multiple social identities, by examining how it relates to one of these social
identities - race/ethnicity – in a pre-adolescent sample. Given the personal benefits of familial
ethnic socialization and ethnic identity affirmation to children from marginalized groups
(Umaña-Taylor et al., 2014; M. A. Zimmerman et al., 2013), it is valuable that these facets of
ethnic identity also seem to predict prosocial attitudes. The observed interaction between ethnic
socialization and SIC suggests that perhaps there are some forms of ethnic socialization which
22
encourage complex social identities and prosocial attitudes and other forms of socialization
which encourage prosocial attitudes independent of SIC. For ethnic minority youth, SIC may not
be necessary for prosocial attitudes, and strong socialization in the absence of SIC may not cause
greater social distancing from racial and ethnic outgroups in majority-minority populations. With
this preliminary evidence that strong socialization and social identity complexity are optimal,
further research which clarifies the function of social identity complexity in demographically
diverse samples of children is warranted. This research could inform the development of
interventions which socialize both the affirmation of children’s ethnic/racial ingroup and wider
acceptance of outgroups through social identity complexity.
23
References
Aboud, F. E. (1989). Children and prejudice. B. Blackwell.
Aboud, F. E., & Amato, M. (2001). Developmental and Socialization Influences on Intergroup
Bias. In R. Brown & S. Gaertner (Eds.), Blackwell handbook of social psychology:
Intergroup Processes (pp. 65–85). Malden: Blackwell Publishers.
Bigler, R. S., & Liben, L. S. (1992). Cognitive Mechanisms in Children’s Gender Stereotyping:
Theoretical and Educational Implications of a Cognitive-Based Intervention. Child
Development, 63(6), 1351. https://doi.org/10.2307/1131561
Brewer, M. B., Gonsalkorale, K., & van Dommelen, A. (2012). Social identity complexity:
Comparing majority and minority ethnic group members in a multicultural society. Group
Processes & Intergroup Relations, 16(5), 529–544.
https://doi.org/10.1177/1368430212468622
Brewer, M. B., & Pierce, K. P. (2005). Social identity complexity and outgroup tolerance.
Personality and Social Psychology Bulletin, 31(3), 428–437.
https://doi.org/10.1177/0146167204271710
Brown, C. S., & Chu, H. (2012). Discrimination, Ethnic Identity, and Academic Outcomes of
Mexican Immigrant Children: The Importance of School Context. Child Development,
83(5), 1477–1485.
Costabile, K. A., & Austin, A. B. (2017). A riot on campus: The effects of social identity
complexity on emotions and reparative attitudes after ingroup-perpetrated violence.
Aggressive Behavior, (June), 1–10. https://doi.org/10.1002/ab.21723
Evans, D. W. (1994). Self-complexity and its relation to development, symptomatology and self-
perception during adolescence. Child Psychiatry and Human Development, 24(3), 173–182.
24
https://doi.org/10.1007/BF02353194
Geldhof, G. J., Bowers, E. P., Boyd, M. J., Mueller, M. K., Napolitano, C. M., Schmid, K. L., …
Lerner, R. M. (2014). Creation of short and very short measures of the five Cs of positive
youth development. Journal of Research on Adolescence, 24(1), 163–176.
https://doi.org/10.1111/jora.12039
Geldhof, G. J., Bowers, E. P., Mueller, M. K., Napolitano, C. M., Callina, K. S., & Lerner, R. M.
(2014). Longitudinal Analysis of a Very Short Measure of Positive Youth Development.
Journal of Youth and Adolescence, 43(6), 933–949. https://doi.org/10.1007/s10964-014-
0093-z
Grant, F., & Hogg, M. A. (2012). Self-uncertainty, social identity prominence and group
identification. Journal of Experimental Social Psychology, 48, 538–542.
https://doi.org/10.1016/j.jesp.2011.11.006
Guerra, R., Rebelo, M., Monteiro, M. B., Riek, B. M., Mania, E. W., Gaertner, S. L., & Dovidio,
J. F. (2010). How should intergroup contact be structured to reduce bias among majority
and minority group children? Group Processes & Intergroup Relations, 13(4), 445–460.
https://doi.org/10.1177/1368430209355651
Hitti, A., Mulvey, K. L., & Killen, M. (2017). Minority and Majority Children’s Evaluations of
Social Exclusion in Intergroup Contexts. In Handbook on Positive Development of Minority
Children and Youth (pp. 281–293). Cham: Springer International Publishing.
https://doi.org/10.1007/978-3-319-43645-6_17
Jugert, P., Noack, P., & Rutland, A. (2011). Friendship Preferences Among German and Turkish
Preadolescents. Child Development, 82(3), 812–829. https://doi.org/10.1111/j.1467-
8624.2010.01528.x
25
Killen, M., & Smetana, J. G. (2010). Future directions: Social development in the context of
social justice. Social Development, 19(3), 642–657. https://doi.org/10.1111/j.1467-
9507.2009.00548.x
Knifsend, C. A., & Juvonen, J. (2013). The role of social identity complexity in inter-group
attitudes among young adolescents. Social Development, 22(3), 623–640.
https://doi.org/10.1111/j.1467-9507.2012.00672.x
Knifsend, C. A., & Juvonen, J. (2014). Social Identity Complexity, Cross-Ethnic Friendships,
and Intergroup Attitudes in Urban Middle Schools. Child Development, 85(2), 709–721.
https://doi.org/10.1111/cdev.12157
Krumpal, I. (2013). Determinants of social desirability bias in sensitive surveys: a literature
review, 2025–2047. https://doi.org/10.1007/s11135-011-9640-9
Lerner, R. M., Lerner, J. V., Almerigi, J. B., Theokas, C., Phelps, E., Gestsdottir, S., … von Eye,
A. (2005). Positive Youth Development, Participation in Community Youth Development
Programs, and Community Contributions of Fifth-Grade Adolescents. The Journal of Early
Adolescence, 25(1), 17–71. https://doi.org/10.1177/0272431604272461
Linville, P. W. (1985). Self-Complexity and Affective Extremity: Don’t Put All of Your Eggs in
One Cognitive Basket. Social Cognition, 3(1), 94–120.
https://doi.org/10.1521/soco.1985.3.1.94
Miller, K. P., Brewer, M. B., & Arbuckle, N. L. (2009). Social Identity Complexity: Its
Correlates and Antecedents. Group Processes & Intergroup Relations, 12(1), 79–94.
https://doi.org/10.1177/1368430208098778
Miller, P. H., Baxter, S. D., Royer, J. A., Hitchcock, D. B., Smith, A. F., Collins, K. L., …
Finney, C. J. (2015). Children’s Social Desirability: Effects of Test Assessment Mode.
26
https://doi.org/10.1016/j.paid.2015.03.039
Pfeifer, J. H., Spears Brown, C., & Juvonen, J. (2007). Teaching tolerance in schools: Lessons
learned since Brown v. Board of Education about the development and reduction of
children’s prejudice. In Social Policy Report (Vol. 21, No. 2, pp. 3–23). Ann Arbor, MI:
Society for Research in Child Development.
Phinney, J. S. (1992). The Multigroup Ethnic Identity Measure. Journal of Adolescent Research,
7(2), 156–176. https://doi.org/10.1177/074355489272003
Rivas-Drake, D., Seaton, E. K., Markstrom, C., Quintana, S., Syed, M., Lee, R. M., … Yip, T.
(2014). Ethnic and Racial Identity in Adolescence: Implications for Psychosocial,
Academic, and Health Outcomes. Child Development, 85(1), 40–57.
https://doi.org/10.1111/cdev.12200
Roccas, S., & Brewer, M. B. (2002). Social identity complexity. Personality and Social
Psychology Review, 6(2), 88–106. https://doi.org/10.1207/s15327957pspr0602_01
Rubin, K. H., Bukowski, W. M., & Laursen, B. P. (2009). Handbook of peer interactions,
relationships, and groups. Guilford Press.
Schmid, K., Hewstone, M., Tausch, N., Cairns, E., & Hughes, J. (2009). Antecedents and
Consequences of Social Identity Complexity: Intergroup Contact, Distinctiveness Threat,
and Outgroup Attitudes. Personality and Social Psychology Bulletin, 35(8), 1085–1098.
https://doi.org/10.1177/0146167209337037
Seaton, E. K., Upton, R., Gilbert, A., & Volpe, V. (2014). A Moderated Mediation Model: Racial
Discrimination, Coping Strategies, and Racial Identity Among Black Adolescents. Child
Development, 85(3), 882–890. https://doi.org/10.1111/cdev.12122
27
Umaña-Taylor, A. J., Quintana, S. M., Lee, R. M., Cross, W. E., Rivas-Drake, D., Schwartz, S.
J., … Sellers, R. M. (2014). Ethnic and Racial Identity During Adolescence and Into Young
Adulthood: An Integrated Conceptualization. Child Development, 85(1), 21–39.
https://doi.org/10.1111/cdev.12196
Umaña-Taylor, A. J., Yazedjian, A., & Bámaca-Gómez, M. (2004). Developing the Ethnic
Identity Scale Using Eriksonian and Social Identity Perspectives. Identity, 4(1), 9–38.
https://doi.org/10.1207/S1532706XID0401_2
Umaña-Taylor, A. J., Zeiders, K. H., Updegraff, K. A., Shanahan, L., Thayer, S., Wheeler, L., …
Rodriguez, S. A. (2013). Family Ethnic Socialization and Ethnic Identity: A Family-Driven
, Youth-Driven , or Reciprocal Process?, 27(1), 137–146. https://doi.org/10.1037/a0031105
Worrell, F. C., Andretta, J. R., & Woodland, M. H. (2014). Cross Racial Identity Scale (CRIS)
scores and profiles in African American adolescents involved with the Juvenile justice
system. Journal of Counseling Psychology, 61(4), 570–580.
https://doi.org/10.1037/cou0000041
Zimmerman, M. A., Stoddard, S. A., Eisman, A. B., Caldwell, C. H., Aiyer, S. M., & Miller, A.
(2013). Adolescent Resilience: Promotive Factors That Inform Prevention. Child
Development Perspectives, 7(4), 215–220. https://doi.org/10.1111/cdep.12042
Zimmerman, S. M., Phelps, E., & Lerner, R. M. (2008). Research in Human Development
Positive and Negative Developmental Trajectories in U . S . Adolescents: Where the
Positive Youth Development Perspective Meets the Deficit Model, 7609.
https://doi.org/10.1080/15427600802274001